ISM9077: How a "Target Y" Inhibitor Just Walked Insilico Into Ophthalmology

Another novel aging + disease program from Insilico has reached developmental candidate stage. I have written some version of that line many times now, and I still do not take it for granted. Each time it lands in my drafts folder, it means the machinery we spent years assembling did the one thing it was built to do — turn a question about biology into a real molecule. Today the sentence carries something extra, something I would not have predicted a couple of years ago: it is opening a door into a disease area we had never worked in before.

The candidate is ISM9077, a potential first-in-class inhibitor of a target I will refer to throughout as "Target Y" for competitive and intellectual-property reasons. We have nominated it as a Preclinical Candidate (PCC) for ocular diseases — with an initial focus on dry age-related macular degeneration (dry AMD), uveitis, and dry eye disease. ISM9077 is our 7th Development Candidate of 2026, and the 32nd PCC we have nominated since 2021. Long-time readers know I do not put those counters at the top of these posts to brag. I put them there because they are the closest thing I have to an honest instrument for measuring whether generative AI drug discovery works as a repeatable process rather than a one-time headline. Thirty-two times, our platform has taken a biological hypothesis and produced a physical compound ready for the next stage. That is not serendipity. It is throughput.

And yet this program is not quite like the thirty-one that came before it — which is exactly why I wanted to write about it honestly rather than let it disappear into a press release.

Why This One Is Genuinely New Territory

Until this nomination, ophthalmology was not part of Insilico's map. It is now. That is a meaningful sentence for a company that has spent the better part of a decade building what I call a dual-purpose pipeline: programs aimed at a specific disease that also engage biology sitting inside the aging process itself. Before ISM9077, that pipeline already stretched across central nervous system disease, fibrosis, cardiovascular disease, metabolism, pain, cancer, and inflammatory bowel disease. Add ophthalmology, and we now have a dual-purpose presence across nearly every major axis of age-related morbidity. For a platform company, being able to say that at all is the whole point.

The commercial backdrop is not trivial either. The global ocular-disease market exceeded USD 40 billion in 2025, and outside forecasts put it on a path to roughly USD 71.58 billion by 2034. Uveitis by itself — one of our three initial indications — accounts for more than 10% of severe visual handicaps in the United States. None of these are boutique problems. They are large, they are underserved, and many of them are still managed with approaches that have barely moved in years.

Here is the part that genuinely excites me as a scientist, and the part where I have to be most disciplined about not overselling: Target Y had not previously been implicated in dry AMD or dry eye disease. Both are authentically novel indications for this biology. No one had drawn a line from this mechanism to these two diseases until our team did the work to find the connection and stand behind it. That is the flavor of novelty I have argued for my entire career — not different for the sake of being different, but different because it opens a door nobody else was even standing near.

The Science: One Molecule Designed to Do Several Jobs

ISM9077 is what we describe as a dual-purpose, potential first-in-class "pipeline-in-a-drug" candidate — a single molecule engineered to act across several related conditions through one shared mechanism, instead of demanding a separate compound for every indication. It was conceived, scored, and refined inside Chemistry42, our generative chemistry platform. The workflow will feel familiar to regular readers, though it still strikes me as remarkable every time: structure-based drug design anchored in our own structural understanding of the target, de novo generative design to reach into chemical space human chemists would rarely think to explore, and an AI-driven, target-specific activity-prediction model built to rank and prioritize what the generators propose. Cycle after cycle of generation and optimization produced ISM9077 — a novel structure against a novel mechanism, with a favorable patent position and a chemical identity no one else has claimed.

What emerged performed, in preclinical models, beyond what I expected across three separate ocular indications:

• Dry AMD models — ISM9077 measurably improved retinal structural integrity and visual function, with efficacy on key endpoints roughly three times that of a current therapy, along with superior histopathological improvement and gains on OCT retinal-thickness and ERG B-wave readouts.
• Uveitis models — the compound lowered ocular inflammation, reduced the release of inflammatory cytokines, and restored retinal function.
• Dry eye models — it increased tear production and quieted corneal inflammation with a fast onset, outperforming Cyclosporine A (CsA), today's standard of care.

Alongside that efficacy, ISM9077 showed a favorable preclinical safety profile with a wide safety margin. I want to keep the qualifier in plain sight rather than bury it: these are preliminary, non-GLP findings, not a guarantee of anything. The molecule also has good permeability and notably strong retinal tissue exposure, reaching 2 to 5.5 times plasma levels — precisely the profile you want if the goal is a molecule that actually acts once it arrives at the eye rather than merely circulating past it. Paired with favorable oral bioavailability, that opens two development routes at once: an oral program and a topical eye-drop formulation. If both hold up, it would mark a real break from the frequent intravitreal injections that dominate advanced ocular-disease care today — a genuinely gentler experience for patients, if we can get there.

And then there is the thread that ties this program to why I care about it beyond its market: Target Y is strongly implicated in aging, longevity, and the inflammatory biology underlying many age-related conditions. That is the dual-purpose thesis made concrete — a disease target that also lives inside the machinery of why we grow old. It is the idea I keep circling back to in this newsletter, because I believe it is the most important structural insight in modern drug discovery: the pathways that drive specific chronic diseases are so often the same pathways that drive aging, that a platform capable of drugging one frequently gets a shot at the other for free.

In Our Own Words

"ISM9077 is our 32nd preclinical candidate since 2021, which makes it a concrete example of generative AI delivering at scale rather than in isolated flashes. It brings together a novel target, a molecule designed from scratch, and a novel formulation possibility in the form of eye drops. Beyond the prospect of a non-invasive, highly effective option for the millions of people living with blinding ocular disease, it deepens our dual-purpose strategy — treating a specific disease while going after aging and healthspan at the same time — precisely because Target Y is so tightly bound up with aging, longevity, and multiple age-related conditions." — Alex Zhavoronkov, PhD

My colleague Feng Ren, our Co-CEO and Chief Scientific Officer, framed the unmet-need side well. His point, as I understood it: ocular disease genuinely needs new options, because existing therapies have never quite balanced toxicity, efficacy, and adherence at the same time. ISM9077 tries to relieve some of those tensions through AI-driven structural design — achieving strong target exposure and the potential for an eye-drop formulation. Dry AMD remains a leading cause of irreversible central vision loss in older adults, and the efficacy we observed in our therapeutic studies, including in Cynomolgus monkey models, represents to our knowledge the first indication of this target's therapeutic value for dry AMD in a system much closer to human biology. Like him, I am eager to see what clinical translation reveals.

The Bigger Platform Story

Step back and ISM9077 is, at bottom, a story about speed. Traditional early discovery often takes 2.5 to 4 years to reach a preclinical candidate. Across our programs, we have repeatedly hit PCC nomination in an average of just 12 to 18 months, synthesizing and testing only 60 to 200 molecules per program on the way. That is not a fluke attached to one fortunate project; it is the sustained output of an integrated system — PandaOmics for target identification and disease-biology insight, Chemistry42 for generative molecular design, and the rest of Pharma.AI shepherding candidates through the translational work that ultimately decides whether preclinical promise survives contact with real biology. Since 2021 that system has yielded 32 PCCs, of which 13 have gone on to earn IND approval or clearance. ISM9077 is simply the newest point on that curve — and the first to plant a flag in ophthalmology.

I do not yet know whether Target Y will turn out to be a target that defines a decade. What I do know is that we could not have found it, designed a molecule against it, and carried it through three distinct disease models on this timeline without the generative AI infrastructure we have spent years building. Whether or not this particular candidate ultimately clears the far higher bar of human proof of concept, the fact that we can make a bet this novel, this deliberately, on this schedule — and then do it again the following quarter — is itself the point. That is what a working drug-discovery platform looks like in 2026: not one lucky molecule, but a system for placing calculated, honest, sometimes uncomfortable bets on biology no one else has dared to try.

Note that this article was prepared using generative AI and may contain hallucinations, inaccuracies, and errors. Please do your own research. While the author is the CEO of Insilico Medicine, the statements and views presented in Forever.ai do not represent the views and opinions of Insilico Medicine.